Detecting device for road monitoring

• US 6,137,531 A
• Filed: 09/08/1997
• Issued: 10/24/2000
• Est. Priority Date: 04/15/1997
• Status: Expired due to Fees

• Current Assignee: Fujitsu Limited
• Original Assignee: Fujitsu Limited
• Inventors: Tofuku, Isao, Haruyama, Hiroshi, Okamoto, Mikihiro, Kanzaki, Yoshiharu
• Primary Examiner(s): Kelley, Chris S.
• Assistant Examiner(s): PHILIPPE, GIMS S
• Application Number:
  US08/925,938
• Time in Patent Office: 1,142 Days
• Field of Search: 348/149, 348/144, 348/143, 348/148, 348/154, 348/155, 348/161, 348/169, 348/131, 348/134, 348/118, 382/103, 382/104, 340/905, 340/601, 701/214, 701/117, 345/431
• US Class Current: 348/149
• CPC Class Codes: B60K 31/0008 including means for detecting potential obstacles in vehicle path
• CPC Class Codes: B60W 2554/00 Input parameters relating to objects
• CPC Class Codes: G01D 21/02 Measuring two or more variables by means not covered by a single other subclass
• CPC Class Codes: G06T 2207/10016 Video; Image sequence
• CPC Class Codes: G06T 2207/10024 Color image
• CPC Class Codes: G06T 2207/30236 Traffic on road, railway or crossing
• CPC Class Codes: G06T 7/254 involving subtraction of images

Abstract

A detecting device for road monitoring which detects traveling states of vehicles on a road (traffic stagnation, stopping of vehicles, etc.) or fallen objects on a road is provided. Luminance change detecting unit prepares background image data by taking a time average of luminance for each picture element, and detects image portions where inputted image data is different in luminance from the prepared background image data. Hue change detecting unit prepares background image data by taking a time average of hue for each picture element, and detects image portions where inputted image data is different in hue from the prepared background image data. Overlapping portion extracting and judging unit extracts overlapping portions between image portions detected by the luminance change detecting unit when it is in the daytime and illuminance is larger than a predetermined value and image portions detected by the hue change detecting unit at the same point of time, and judges the extracted overlapping portions to represent objects to be detected. Comprehensive portion extracting and judging unit extracts comprehensive portions covering image portions detected by the luminance change detecting unit when it is in the daytime and illuminance is smaller than a predetermined value and image portions detected by the hue change detecting unit at the same point of time, and judges the extracted comprehensive portions to represent objects to be detected.

Claims

1. A detecting device for road monitoring which detects traveling states of vehicles on a road or fallen objects on a road, comprising:

a color image pick-up camera for picking up images of a road;

luminance change detecting means for detecting, in image data fed from said color image pick-up camera, image portions where the image data is different in luminance from background image data which is prepared by taking a time average of luminance for each picture element;

hue change detecting means for detecting, in image data fed from said color image pick-up camera, image portions where the image data is different in hue from background image data which is prepared by taking a time average of hue for each picture element; and

overlapping portion extracting and judging means for extracting overlapping portions between image portions detected by said luminance change detecting means when it is in the daytime and illuminance is larger than a predetermined value and image portions detected by said hue change detecting means at the same point of time, and judging the extracted overlapping portions to represent objects existing on the road and to be detected.

2. A detecting device for road monitoring which detects traveling states of vehicles on a road or fallen objects on a road, comprising:

a color image pick-up camera for picking up images of a road;

luminance change detecting means for detecting, in image data fed from said color image pick-up camera, image portions where the image data is different in luminance from background image data which is prepared by taking a time average of luminance for each picture element;

hue change detecting means for detecting, in image data fed from said color image pick-up camera, image portions where the image data is different in hue from background image data which is prepared by taking a time average of hue for each picture element; and

comprehensive portion extracting and judging means for extracting comprehensive portions covering image portions detected by said luminance change detecting means when it is in the daytime and illuminance is smaller than a predetermined value and image portions detected by said hue change detecting means at the same point of time, and judging the extracted comprehensive portions to represent objects existing on the road and to be detected.

3. A detecting device for road monitoring which detects traveling states of vehicles on a road or fallen objects on a road, comprising:

a color image pick-up camera for picking up images of a road;

designated-color image portion detecting means for detecting, in image data fed from said color image pick-up camera, image portions having a designated color;

color intensity change detecting means for detecting, within the image portions detected by said designated-color image portion detecting means, image portions where the image data is different in color intensity from background image data which is prepared by taking a time average of color intensity for each picture element; and

judging means for judging the image portions detected by said color intensity change detecting means in the nighttime to represent objects existing on the road and to be detected.

4. A detecting device for road monitoring according to claim 3, wherein said detecting device further comprises excluding means for detecting, in the image data fed from said color image pick-up camera, picture elements having spectrum of all colors and excluding the detected picture elements from the image portions detected by said designated-color image portion detecting means.

5. A detecting device for road monitoring according to claim 3, wherein said designated-color image portion detecting means detects, in the image data fed from said color image pick-up camera, image portions having a color of taillights of vehicles.

6. A detecting device for road monitoring according to claim 5, wherein said detecting device further comprises image processing means for expanding and contracting the image portions detected by said color intensity change detecting means in a direction transversal to a road represented in the image data, thereby making image portions representing a pair of taillights of a vehicle into one image.

7. A detecting device for road monitoring which detects traveling states of vehicles on a road or fallen objects on a road, comprising:

a color image pick-up camera for picking up images of a road;

dividing means for dividing image data fed from said color image pick-up camera into a plurality of image sections according to lanes of road represented in the image data, so that each lane represented in the image data consists of a plurality of image sections being in line;

average luminance calculating means for calculating a spatial average of luminance in each image section prepared by said dividing means;

threshold value determining means for automatically determining a threshold value for each image section, based on the spatial average of luminance in each image section calculated by said average luminance calculating means;

background image data preparing means for obtaining a time average of luminance for each picture element in each image section prepared by said dividing means, thereby preparing background image data for each image section;

difference calculating means for calculating difference in luminance for each picture element between each image section prepared by said dividing means and the background image data for each image section prepared by said background image data preparing means;

binary processing means for making the difference in luminance calculated by said difference calculating means into binary data, using the threshold value determined for each image section by said threshold value determining means; and

object detecting means for detecting objects existing on the road, based on the binary data prepared by said binary processing means.

8. A detecting device for road monitoring which detects traveling states of vehicles on a road or fallen objects on a road, comprising:

a color image pick-up camera for picking up images of a road;

dividing means for dividing image data fed from said color image pick-up camera into a plurality of image sections according to lanes of road represented in the image data, so that each lane represented in the image data consists of a plurality of image sections being in line;

average hue calculating means for calculating a spatial average of hue in each image section prepared by said dividing means;

threshold value determining means for determining a threshold value for each image section, based on the spatial average of hue in each image section calculated by said average hue calculating means;

background image data preparing means for obtaining a time average of hue for each picture element in each image section prepared by said dividing means, thereby preparing background image data for each image section;

difference calculating means for calculating difference in hue for each picture element between each image section prepared by said dividing means and the background image data for each image section prepared by said background image data preparing means;

binary processing means for making the difference in hue calculated by said difference calculating means into binary data, using the threshold value determined for each image section by said threshold value determining means; and

object detecting means for detecting objects existing on the road, based on the binary data prepared by said binary processing means.